Greg Pfeil

@juk0de I don’t think you need to do quite what I did. [I commented on your issue](https://github.com/ryantm/agenix/issues/305#issuecomment-3395433809), with what I think will work for you.

> Yes, I agreed that tests should be considered referenced, though my first reaction was that it sucks to special-case them. But maybe it's not. If you’re special-casing them anyway,...

I have an M1 MacBook, not yet updated to Sequoia, and I started having a similar problem just yesterday. I _have_ had flakiness in the past, where disabling the internal...

The code has changed a bit, but the failure is still similar. I updated the transcript to a working command (and one that won’t pull base) ````markdown ```ucm > help...

The transcript here is no longer invalid (as of #5826), so no longer gives an error.

Discussed #6679 (and the comment just above) with @str4d and we came up with the following structure to be used within each spending authority: - available - confirmed - unconfirmed...

> > 1. “available” is not called “spendable” because we use the same structure for watchonly > > Could watch-only balances be merged into the existing structure, to get rid...

Did you run into this issue outside of a transcript originally? I can replicate the behavior in a transcript, but as soon as I try to minimize the test case...

Ah, I can replicate it in a scratch file too. But still odd that unit tests, etc. work.

Yeah, it would be nice if we could cache incrementally, so then repeated builds would make progress. For the time being, I’ve relied on pushing to the cache manually after...